INDEX

Explanations

words related to political conflicts and current events

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs3072-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

3,072

Data Type

torch.float32

Hook Point

blocks.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Os

-0.83

 Indigo

-0.76

 Roose

-0.75

 Niet

-0.75

 Winds

-0.72

 Horton

-0.70

 Aware

-0.69

tions

-0.67

 Lutheran

-0.67

POSITIVE LOGITS

 blackmail

0.94

 augment

0.92

 pretext

0.85

 efficiently

0.83

 supplement

0.82

 enhance

0.82

 shortcut

0.80

 excuse

0.79

 justify

0.79

 coping

0.79

Activations Density 1.233%

words related to political conflicts and current events

No Comments

No Known Activations